WO2020042500A1 - 一种基于远程眼底筛查的健康大数据服务方法和*** - Google Patents

一种基于远程眼底筛查的健康大数据服务方法和*** Download PDF

Info

Publication number
WO2020042500A1
WO2020042500A1 PCT/CN2018/124488 CN2018124488W WO2020042500A1 WO 2020042500 A1 WO2020042500 A1 WO 2020042500A1 CN 2018124488 W CN2018124488 W CN 2018124488W WO 2020042500 A1 WO2020042500 A1 WO 2020042500A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
analyzed
fundus image
fundus
qualified
Prior art date
Application number
PCT/CN2018/124488
Other languages
English (en)
French (fr)
Inventor
余轮
欧霖杰
王丽纳
林嘉雯
邱应强
薛岚燕
曹新容
Original Assignee
福州依影健康科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 福州依影健康科技有限公司 filed Critical 福州依影健康科技有限公司
Priority to US17/271,611 priority Critical patent/US20210391056A1/en
Publication of WO2020042500A1 publication Critical patent/WO2020042500A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H30/00—ICT specially adapted for the handling or processing of medical images
    • G16H30/20—ICT specially adapted for the handling or processing of medical images for handling medical images, e.g. DICOM, HL7 or PACS
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B3/00—Apparatus for testing the eyes; Instruments for examining the eyes
    • A61B3/0016—Operational features thereof
    • A61B3/0025—Operational features thereof characterised by electronic signal processing, e.g. eye models
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B3/00—Apparatus for testing the eyes; Instruments for examining the eyes
    • A61B3/0016—Operational features thereof
    • A61B3/0041—Operational features thereof characterised by display arrangements
    • A61B3/0058—Operational features thereof characterised by display arrangements for multiple images
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B3/00—Apparatus for testing the eyes; Instruments for examining the eyes
    • A61B3/10—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions
    • A61B3/12—Objective types, i.e. instruments for examining the eyes independent of the patients' perceptions or reactions for looking at the eye fundus, e.g. ophthalmoscopes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/64—Analysis of geometric attributes of convexity or concavity
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/66—Analysis of geometric attributes of image moments or centre of gravity
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H30/00—ICT specially adapted for the handling or processing of medical images
    • G16H30/40—ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20081—Training; Learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20084—Artificial neural networks [ANN]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30041—Eye; Retina; Ophthalmic
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30101—Blood vessel; Artery; Vein; Vascular

Definitions

  • the invention relates to the field of health big data, in particular to a method and system for health big data service based on remote fundus screening.
  • Diabetic Retinopathy has affected More than 100 million people; according to the number and structure of the population, China currently has about 250 million hypertensive patients, 1 in 3 adults with hypertension, accounting for about 1/3 of the total number of people with hypertension in the world The prevalence rate is on the rise and rises with age, while the control rate is only 5.7%. China has more than 110 million diabetics. Diabetes and its complications cause a serious socioeconomic burden.
  • the fundus camera technology for diabetic retinopathy (DR) screening has matured today, due to various types of fundus cameras and their different working modes, the size, resolution, and structure of the fundus images obtained not exactly. Even if the same user has the same eye and is collected by different devices or at different times, the fundus images obtained may not be able to achieve individual comparisons of multiple inspection images and quantitative quantitative analysis due to different perspectives and resolutions. It is also difficult to perform quantitative analysis, statistics, or comparison of the retinal pathological characteristics, location, size, or vascular changes of fundus images collected by different people or the same person at different times or on different devices, affecting the application and health of their structured data. Data acquisition, establishment, update and comparison.
  • a method for health big data service based on remote fundus screening comprising the steps of obtaining information to be analyzed sent by a remote terminal agency, the information to be analyzed includes: fundus images and personal data; pre-judging the information to be analyzed, Judging whether the information to be analyzed is qualified; if the information to be analyzed is qualified, extract characteristic data from the information to be analyzed and form a structured quantitative index; arrange the characteristic data and the quantitative index according to a knowledge calculation model And analysis to obtain an analysis conclusion; and storing the information to be analyzed, characteristic data, quantitative indicators, and analysis conclusion in the pre-designed database.
  • the "pre-judging the information to be analyzed to determine whether the information to be analyzed is qualified” further includes the step of: the pre-judging includes: whether the fundus image is indeed a fundus image, and the fundus image Whether the structure is complete, whether the fundus image is clear, and whether the fundus image is available; if the information to be analyzed is qualified, the relevant qualified information is returned to the remote terminal agency; if the information to be analyzed is not qualified, it is returned The relevant non-conformance information is sent to the remote terminal organization, and the related non-conformity information is used to prompt the remote terminal organization to collect the information to be analyzed again.
  • the step of “pre-judging the information to be analyzed and determining whether the information to be analyzed is qualified” further includes the step of: according to a preset rule, before returning the pre-judgment result to the remote terminal organization, the remote terminal The agency sends a prompt message to prompt the user not to leave the remote terminal agency until the prompt message that the information to be analyzed is qualified is returned.
  • the step of “pre-judging the information to be analyzed and determining whether the information to be analyzed is qualified” further includes the steps of: if the information to be analyzed is qualified, returning relevant qualified information to a remote terminal organization; and the remote terminal organization Acquiring the relevant qualified information, and notifying the user whether to continue to wait until the analysis conclusion is reached according to a preset rule.
  • the "whether the fundus image structure is complete" further includes the steps of identifying and calibrating the optic disc and the macula of the fundus image, and determining whether the fundus image includes the optic disc and the macula according to the recognition result.
  • the fundus image includes the optic disc and the macula, and it is determined whether the optic disc and the macula are within a preset interval of the fundus image according to the calibration result. If the optic disc and the macula are within a preset interval of the fundus image, the fundus image The structure is complete.
  • the step of “extracting characteristic data from the fundus image and forming a structured quantified index” further includes the step of calculating a quantization parameter of the temporal side of the optic disc and the fovea of the macula according to the calibrated optic disc and the macula.
  • a health big data service system based on remote fundus screening includes: a fundus image acquisition module and a remote analysis center module; the fundus image acquisition module is connected to the remote analysis center module; and the fundus image acquisition module is used to:
  • the information to be analyzed includes: fundus images and personal data, and sends the information to be analyzed to a remote analysis center module; the remote analysis center module is configured to: receive the information to be analyzed, and The information to be analyzed is pre-judgmented to determine whether the information to be analyzed is qualified; if the information to be analyzed is qualified, characteristic data is extracted from the information to be analyzed and a structured quantitative index is formed; The characteristic data and the quantitative index are sorted and analyzed to obtain an analysis conclusion; the information to be analyzed, the characteristic data, the quantitative index, and the analysis conclusion are stored in the pre-designed database.
  • the pre-judgment reading includes one or more of whether the fundus image is indeed a fundus image, whether the fundus image structure is complete, whether the fundus image is clear, and whether the fundus image is available;
  • the remote analysis center module is further configured to: if the information to be analyzed is qualified, return the relevant qualified information to the fundus image acquisition module; if the information to be analyzed is not qualified, return the relevant unqualified information to the fundus image acquisition module, the relevant unqualified information It is used to prompt: the fundus image acquisition module reacquires the qualified information to be analyzed.
  • the fundus image acquisition module is further configured to: according to a preset rule, before returning the pre-reading result to the fundus image acquisition module, issue a prompt message to prompt the user not to leave the fundus image acquisition module until the returned information to be analyzed is qualified. Prompt message.
  • the remote analysis center module is further configured to: if the information to be analyzed is qualified, return relevant qualified information to the fundus image acquisition module; the fundus image acquisition module is further configured to: obtain the relevant qualified information, and according to a preset The rules inform the user if they need to continue to wait until the analysis is concluded.
  • the beneficial effect of the present invention is: by acquiring the information to be analyzed sent by the remote terminal mechanism, the information to be checked includes: fundus images and personal data, pre-judging and reading the information to be analyzed, and determining whether the information to be analyzed is qualified, It is very important to form a complete closed-loop quality assurance system, so that the system can make every piece of information to be analyzed fully available, ensuring reliable user information acquisition, enhancing the user experience, and It helps to form a big data information database that can be analyzed and updated; if the information to be analyzed is qualified, feature data is extracted from the information to be analyzed and a structured quantitative index is formed; the feature data and the Quantitative indicators are stored in a pre-designed database; the characteristic data and quantitative indicators are sorted and analyzed according to the knowledge calculation model to obtain analysis conclusions; the information to be analyzed, characteristic data, quantitative indicators and analysis conclusions are stored in the Pre-designed database.
  • pre-judgment reading of the information to be analyzed can be guaranteed to be 100% available until the last to extract feature data.
  • the remote analysis center finds that the information to be analyzed is not available later, the user It has to run again, giving users a bad experience and wasting users' time.
  • the available information to be analyzed not only ensures the stability and accuracy of the diagnosis results, but also improves the diagnosis efficiency. Avoid useless rework time.
  • the remote terminal organization can be instructed to not allow the user to leave temporarily until the prompt message that the information to be analyzed is qualified is returned before leaving. The entire process avoids the unqualified information to be analyzed. However, the situation that the user has left has caused a bad experience for the user.
  • the fundus image is qualified, feature data is extracted from the fundus image, and a structured quantization index is formed. That is, according to the calibrated optic disc and the macula, the quantitative parameters of the temporal side of the optic disc and the fovea of the macula are calculated. Because the absolute distance values of the two are almost the same for normal people, the parameters of subsequent quantitative analysis are obtained based on the absolute distance from the optic disc temporal to the macular fovea and the diameter of the optic disc, and the obtained data are expressed from the absolute representation Convert to a relative representation and normalize to form meaningful and comparable data. It is ensured that fundus images from different sources can form meaningful and comparable quantitative indicators, so that all fundus images are basically comparable.
  • FIG. 1 is a flowchart of a method for health big data service based on remote fundus screening according to a specific embodiment
  • FIG. 2 is a module diagram of a health big data service system based on remote fundus screening according to a specific embodiment.
  • a health big data service system based on remote fundus screening
  • a remote analysis center module 202.
  • all or part of the steps in a method for health big data service based on remote fundus screening may be completed by a program instructing related hardware, and the program may be stored in a computer device.
  • the readable storage medium is used to perform all or part of the steps described in the following embodiments.
  • the computer equipment includes, but is not limited to, a personal computer, a server, a general-purpose computer, a dedicated computer, a network device, a smart mobile terminal, a smart home device, a wearable smart device, and the like; the storage medium includes, but is not limited to: RAM , ROM, mobile hard disk, network server storage, network cloud storage, etc.
  • a specific implementation of a health big data service method based on remote fundus screening is as follows:
  • Step S101 Obtain information to be analyzed sent by a remote terminal mechanism, where the information to be analyzed includes: fundus images and personal data.
  • a fundus image acquisition terminal and any computer equipment that can receive information can be provided in the remote terminal mechanism, such as a PC, which collects the fundus images through the fundus image acquisition terminal, and then transmits the fundus images to the PC, and also personal data Input into the PC, and the PC sends the fundus image and personal data to the remote analysis center.
  • the remote terminal mechanism (such as the underlying medical clinic) may collect the fundus image of the user A through the fundus camera, and the fundus camera transmits the fundus image to the computer of the terminal application mechanism through a USB cable, and also Personal information or personal information can be entered into the computer, and the computer sends this information to the remote interpretation center.
  • the personal information or consultation information includes: name, identity card, height, weight, waist circumference, family genetic medical history, medication status, blood glucose, blood pressure, vision status, exercise status, diet status, lifestyle habits, and Whether smoking or drinking one or more of them.
  • step S102 pre-judging and reading the information to be analyzed, and determining whether the information to be analyzed is qualified.
  • the pre-judgment reading includes one or more of whether the fundus image is indeed a fundus image, whether the fundus image structure is complete, whether the fundus image is clear, and whether the fundus image is available.
  • the following methods can be adopted: obtaining the pre-judgment information input by the quality inspector; combining the pre-judgment information input by the quality inspector and the pre-judgment result of automatic analysis of the fundus image to determine whether the fundus image is qualified; the input information includes: fundus image quality level.
  • the fundus image can be automatically analyzed, for example, using an image collected in advance by a professional doctor to classify the quality level, training an SVM model, so that the model can perform image quality level classification. In this way, when a fundus image comes from the grassroots, on the one hand, the trained svm model is used to judge the fundus image, and on the other hand, the quality of the fundus image judged by the quality inspector (such as a professional ophthalmologist) is also obtained.
  • the quality of the fundus image judged by the quality inspector such as a professional ophthalmologist
  • the fundus image analysis can also be performed automatically, without the need for quality control personnel to participate in the check.
  • the received fundus image it can be judged whether the fundus image is indeed the fundus image. If the fundus image is not the fundus image, that is: it may be a prank or a manipulation error when the fundus image is selected for sending by the remote terminal agency staff, etc. If it is caused, the interpretation is directly ended, and the corresponding prompt information is returned to the remote terminal mechanism, informing it that the fundus image acquired is not the fundus image, and the user performs the fundus image acquisition operation accordingly.
  • the fundus image is a fundus image
  • the following methods can be used: identifying and calibrating the optic disc and the macula of the fundus image, and judging whether the fundus image contains Optic disk and macula, if the fundus image includes the optic disk and the macula, determine whether the optic disk and the macula are within a preset interval of the fundus image according to the calibration result, and if the visual disk and the macula are within a preset interval of the fundus image, Then the fundus image structure is complete.
  • the details are as follows: green channel selection, median filtering, limited contrast enhancement, and grayscale normalization are performed on the fundus image to be inspected.
  • the redundant background in the fundus image can be removed and the noise can be effectively denoised, which is more conducive to subsequent fundus image analysis.
  • the details are as follows: In any color fundus image, there is more noise under the blue channel, and useful information is basically lost. The two spots under the red channel are more prominent, and dark blood vessels, microhemangiomas and other information are lost. Therefore, in this implementation In the method, the green fundus image to be inspected is selected for the green channel to retain and highlight the fundus vessels to the greatest extent.
  • the pre-processed fundus image is extracted with a binarized blood vessel map by the Otsu algorithm, and the binarized blood vessel map is corroded by a morphological method to obtain a main blood vessel.
  • the following method can be adopted: the threshold value is calculated through the Otsu algorithm for the pre-processed fundus image, and pixels with a gray value greater than the threshold value are identified as blood vessels according to the following formula;
  • a parabolic fit of the main blood vessel is performed according to the least square method, the parameters of the parabola are determined, and the vertex of the parabola is calculated.
  • a ring mask is constructed and defined as the foveal search range; then In the search area, the positioning of the fovea is completed according to the characteristics of the lowest brightness of the fovea.
  • the position of the fovea is determined using a fast search method based on the brightness contrast between the regions; finally, according to the brightness information, the fovea region is fitted with the fovea as the center and the circle is fitted.
  • the structural integrity of the fundus image was determined. According to the judgment conditions shown in Table 1, the image that satisfies the condition is an image with qualified integrity. Among them, Dod is the diameter of the optic disc.
  • the fundus image is a fundus image
  • the following methods may be used: judging whether the small blood vessels on the surface of the optic disc and the retinal nerve fiber layer of the posterior pole of the fundus image are distinguishable. The small blood vessels on the surface of the optic disc and the nerve fiber layer in the posterior pole of the image can be discerned, and then the fundus image is qualified. details as follows:
  • the optic disc center and macular center depression determine the optic disc as the center of the circle, a region of 1.5 times the diameter of the optic disc as the center of interest1, with the macular center as the center of the circle, and the area of 1 times the diameter of the optic disc as the region of interest 2;
  • Step S103 If the information to be analyzed is qualified, feature data is extracted from the information to be analyzed, and a structured quantitative index is formed.
  • the details are as follows: According to the calibrated optic disc and the macula, the quantitative parameters of the temporal side of the optic disc and the fovea of the macula are calculated. According to the optic disc center coordinates and optic disc radius, calculate the optic disc temporal coordinates (ODX, ODY); based on the optic disc temporal coordinates and macular foveal coordinates, calculate the absolute distance between the optic disc temporal and the macular fovea, and calculate the fundus image according to the following formula The Euclidean distance between the two is taken as the absolute distance between the center of the optic disc and the center of the macula in this image;
  • all coordinate values are based on the upper left pixel of the fundus image as the origin.
  • the concavity in the macula is generally about 3mm from the temporal margin of the optic disc, so according to the absolute distance from the optic disc temporal to the macular fovea and the diameter of the optic disc, the standard d for subsequent quantitative analysis is obtained according to the following formula:
  • the obtained data is converted from an absolute representation to a relative representation, and through this normalization process, meaningful and comparable data is formed.
  • Step S104 According to the knowledge calculation model, sort and analyze the characteristic data and quantified indicators to obtain an analysis conclusion.
  • Step S105 storing the information to be analyzed, the characteristic data, the quantitative index, and the analysis conclusion into the pre-designed database.
  • DR screening Today, fundus camera technology for diabetic retinopathy (DR) screening has matured, and DR screening has prevention and treatment guidelines and diagnostic standards that are related to diabetes and can guide treatment; the eyes are the entire body The only place where blood vessels and nerves can be seen directly without surgery. Medical evidence shows that the circulatory system of the retina and brain has similar anatomical, physiological and embryonic development characteristics. Therefore, the degree of pathological changes in the whole body, especially in the cerebral arteries and small and medium arteries can be understood through the fundus blood vessels. According to the guidelines for the prevention and treatment of hypertension in China, retinal arterial disease can reflect the condition of small vessel disease.
  • “Knowledge Computing Model” provides an eye The statistical, calculation and analysis methods of blood vessel characteristics of the bottom image are very active for timely discovery of characteristic information such as retinal retinopathy and fundus vascular changes, provision of auxiliary diagnostic information or health management and health service recommendations, and development of health big data services. significance.
  • a structured quantitative index and the "knowledge base" of the in-depth professional design are formed, where the quantitative index includes: previous medical history and height and weight , Waist circumference, daily exercise diet habits, history of diabetes and previous treatments, history of hypertension and previous treatments and other personal information or consultation information such as family genetic history, lifestyle habits, microhemangioma, bleeding points, hard exudation Results of DR interpretation related to the number, area, and position of cotton plaque, plus the presence of new blood vessels, macular edema, arteriovenous ratio in the region of interest, narrowing of arterial diameter, and arteriovenous cross pressure
  • the "knowledge calculation model” it is possible to provide a fundus retinopathy and blood vessels based on the "knowledge calculation model”.
  • the statistical, calculation and analysis methods of change and other characteristics namely "the disease early warning and health assessment engine”.
  • the information to be checked includes: fundus images and personal data, pre-judging the information to be analyzed, and determining whether the information to be analyzed is qualified; if the information to be analyzed is qualified Extracting feature data from the information to be analyzed and forming a structured quantitative index; storing the feature data and the quantitative index in a pre-designed database; and performing the feature data and the quantitative index according to a knowledge calculation model Arrange and analyze to obtain the analysis conclusion; store the information to be analyzed, the characteristic data, the quantitative index and the analysis conclusion into the pre-designed database.
  • the available information to be analyzed not only ensures the stability and accuracy of the diagnosis results, but also improves the diagnosis efficiency. Avoid useless rework time.
  • the remote terminal organization can be instructed to not allow the user to leave temporarily until the prompt message that the information to be analyzed is qualified is returned before leaving. The entire process avoids the unqualified information to be analyzed. However, the situation that the user has left has caused a bad experience for the user.
  • the fundus image is qualified, feature data is extracted from the fundus image, and a structured quantization index is formed. That is, according to the calibrated optic disc and the macula, the quantitative parameters of the temporal side of the optic disc and the fovea of the macula are calculated.
  • the parameters of subsequent quantitative analysis are obtained based on the absolute distance from the optic disc temporal to the macular fovea and the diameter of the optic disc, and the obtained data are expressed from the absolute representation Convert to a relative representation and normalize to form meaningful and comparable data. It is ensured that fundus images from different sources can form meaningful and comparable quantitative indicators, so that all fundus images are basically comparable.
  • the positioning of the optic disc and the macula can also be completed manually.
  • the method may further include the step that the remote terminal organization is provided with specific software that can be used to pre-read the fundus image and personal data locally. If the judgment result is qualified, a corresponding prompt message is issued to remind the user that he can leave or continue to rest nearby to wait for the analysis result of the remote analysis center.
  • the step of “pre-judging the information to be analyzed and determining whether the information to be analyzed is qualified” further includes the step of: according to a preset rule, before returning the pre-judgment result to the remote terminal organization, the remote terminal The agency sends a prompt message to prompt the user not to leave the remote terminal agency until the prompt message that the information to be analyzed is qualified is returned.
  • the following method may be adopted: if the remote terminal organization is not provided with software that can be used to pre-read the fundus images and personal data locally, according to a preset rule (that is, whether the remote terminal organization is a protocol service of the remote analysis center) (Or closely related, when it is not necessary to purchase specific software, according to the agreement on the process or quality control system), before returning the pre-reading result to the remote terminal organization, the remote terminal organization sends a prompt message to prompt the user not to leave the remote
  • the terminal mechanism may also allow the user to rest nearby and wait for the remote interpretation consultation result of the remote interpretation consultation center until the prompt information that the information to be analyzed is qualified is returned.
  • the step of “pre-judging the information to be analyzed and determining whether the information to be analyzed is qualified” further includes the steps of: if the information to be analyzed is qualified, returning relevant qualified information to a remote terminal organization; and the remote terminal organization Acquiring the relevant qualified information, and notifying the user whether to continue to wait until the analysis conclusion is reached according to a preset rule.
  • the following method can be adopted: After receiving the prompt that the information to be analyzed is qualified, the remote terminal structure can inform the user whether to continue to wait until the analysis conclusion is reached according to the actual situation.
  • the fundus image acquisition module 201 includes at least: a fundus image acquisition camera and a computer; and the remote analysis center module 202 may be a storage device.
  • a corresponding remote analysis center APP is installed on the storage device, or the corresponding remote analysis center webpage is directly opened, and the to-be-analyzed information transmitted by the fundus image acquisition module 201 can be processed.
  • a specific implementation of a health big data service system 200 based on remote fundus screening is as follows:
  • a health big data service system 200 based on remote fundus screening includes: fundus image acquisition module 201 and remote analysis center module 202; the fundus image acquisition module 201 is connected to the remote analysis center module 202; and the fundus image acquisition Module 201 is configured to: obtain information to be analyzed, the information to be checked includes: fundus images and personal data, and send the information to be analyzed to a remote analysis center module 202; the remote analysis center module 202 is configured to: receive the information The information to be analyzed, and the information to be analyzed is pre-judged to determine whether the information to be analyzed is qualified; if the information to be analyzed is qualified, characteristic data is extracted from the information to be analyzed, and a structured quantitative index is formed; According to the knowledge calculation model, the characteristic data and the quantitative index are sorted and analyzed to obtain an analysis conclusion; the information to be analyzed, the characteristic data, the quantitative index, and the analysis conclusion are stored in the pre-designed database.
  • the pre-judgment reading includes one or more of whether the fundus image is indeed a fundus image, whether the fundus image structure is complete, whether the fundus image is clear, and whether the fundus image is available;
  • the remote analysis center module 202 is further configured to: if the information to be analyzed is qualified, return the relevant qualified information to the fundus image acquisition module 201; if the information to be analyzed is not qualified, return the relevant unqualified information to the fundus image acquisition module 201, where the relevant The unqualified information is used to prompt: the fundus image acquisition module 201 re-acquires the qualified information to be analyzed.
  • the fundus image acquisition module 201 is further configured to: according to a preset rule, before returning the pre-reading result to the fundus image acquisition module 201, send out a prompt message to prompt the user not to leave the fundus image acquisition module 201 until returning to standby. Prompt information that analysis information is qualified.
  • the remote analysis center module 202 is further configured to: if the information to be analyzed is qualified, return relevant qualified information to the fundus image acquisition module 201; the fundus image acquisition module 201 is further configured to: obtain the relevant qualified information, and According to a preset rule, the user is informed whether to continue to wait until the analysis conclusion is reached.
  • a health big data service system 200 based on remote fundus screening obtains information to be analyzed through the fundus image acquisition module 201.
  • the information to be tested includes: fundus images and personal data, and the remote analysis center module 202 pre-judges the information to be analyzed. Judging whether the information to be analyzed is qualified; if the information to be analyzed is qualified, extract characteristic data from the information to be analyzed and form a structured quantitative index; arrange the characteristic data and the quantitative index according to a knowledge calculation model And analysis to obtain an analysis conclusion; and storing the information to be analyzed, characteristic data, quantitative indicators, and analysis conclusion in the pre-designed database.
  • pre-judgment reading of the information to be analyzed can be guaranteed to be 100% available until the last to extract feature data.
  • the remote analysis center finds that the information to be analyzed is not available later, the user It has to run again, giving users a bad experience and wasting users' time.
  • the available information to be analyzed not only ensures the stability and accuracy of the diagnosis results, but also improves the diagnosis efficiency. Avoid useless rework time.
  • the remote terminal organization can be instructed to not allow the user to leave temporarily until the prompt message that the information to be analyzed is qualified is returned before leaving. The entire process avoids the unqualified information to be analyzed. However, the situation that the user has left has caused a bad experience for the user.
  • the fundus image is qualified, feature data is extracted from the fundus image, and a structured quantization index is formed. That is, according to the calibrated optic disc and the macula, the quantitative parameters of the temporal side of the optic disc and the fovea of the macula are calculated. Because the absolute distance values of the two are almost the same for normal people, the parameters of subsequent quantitative analysis are obtained based on the absolute distance from the optic disc temporal to the macular fovea and the diameter of the optic disc, and the obtained data are expressed from the absolute representation Convert to a relative representation and normalize to form meaningful and comparable data. It is ensured that fundus images from different sources can form meaningful and comparable quantitative indicators, so that all fundus images are basically comparable.

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Public Health (AREA)
  • General Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Epidemiology (AREA)
  • Primary Health Care (AREA)
  • Radiology & Medical Imaging (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Pathology (AREA)
  • Databases & Information Systems (AREA)
  • Biophysics (AREA)
  • Ophthalmology & Optometry (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • Veterinary Medicine (AREA)
  • Quality & Reliability (AREA)
  • Geometry (AREA)
  • Signal Processing (AREA)
  • Eye Examination Apparatus (AREA)

Abstract

一种基于远程眼底筛查的健康大数据服务方法与***。基于远程眼底筛查的健康大数据服务方法包括步骤：获取远程终端机构发送的待分析信息（S101）；对待分析信息进行预判读，判断待分析信息是否合格（S102）；若合格，对待分析信息提取特征数据，并形成结构化的量化指标（S103）；根据知识计算模型，对特征数据和量化指标进行整理与分析，获得分析结论（S104）；将待分析信息、特征数据、量化指标和分析结论存放至数据库（S105）。通过以上步骤使得无论是什么型号的眼底相机、什么样的工作模式、最后获得的眼底图像经过处理之后，都可以获得可比性统一的量化指标、特征数据，建立起整个大数据服务平台，将大大地有利于实现个性化的健康大数据服务等。

Description

一种基于远程眼底筛查的健康大数据服务方法和*** 技术领域
本发明涉及健康大数据领域，特别涉及一种基于远程眼底筛查的健康大数据服务方法和***。
背景技术
根据世界卫生组织新近发布的消息，全球范围内有3.47亿人被诊断患有糖尿病，并预计到2040年，这个数字将会超过6.4亿，而估计糖尿病性视网膜病变(Diabtic Retinopathy,DR)已经影响超过1亿人；按人口的数量与结构估算，目前我国约有2.5亿高血压患者，每3个成年人中就有1人患有高血压，约占全球高血压总人数的1/3患病率呈上升趋势，且随年龄增高而上升，而控制率仅为5.7％。中国的糖尿病患者也已经超过1.1亿。糖尿病及其并发症造成严重的社会经济负担。
但是，我国迄今缺乏脑卒中和DR、DN以及青光眼、白内障等重大疾病或重大并发症的高效预警或规模化的筛查平台；移动医疗难以获得脑心眼肾等靶器官的个体化精准信息，难以实现个性化的健康服务！
今天虽然用于糖尿病视网膜病变(DR)筛查的眼底照相机技术已经成熟，然而，由于各种型号的眼底相机及其不同的工作模式，导致所获得的眼底图像的尺寸、分辨率、结构等都不尽相同。即使是同一位用户同一只眼睛，通过不同设备或不同时间的采集，所获得眼底图像也可能因为视角、分辨率的不同，导致无法实现个人多次检查图像的比对和指标性的定量分析，也难以对不同人或同一个人不同时间或不同设备上采集的眼底图象的视网膜病变特征、位置、大小或血管改变情况进行定量分析、统计或比较，影响了其结构化数据的应用和健康大数据的获取、建立、更新与比较。目前，暂未有相关文献注意到这些问题或明确指出如何解决这一问题！因此，面对海量的用户定期筛查与采集的眼底图象，如何在对单张眼底图像中的关键结构、与各类 疾病相关的病灶进行分析的基础上，形成可比较的、有意义的量化指标，从而支持眼底图像的比对分析和统计是一个亟需待解决的问题。
发明内容
为此，需要提供一种基于远程眼底筛查的健康大数据服务方法，用以解决无法对海量眼底图像处理分析形成量化指标，形成结构化数据的问题。具体技术方案如下：
一种基于远程眼底筛查的健康大数据服务方法，包括步骤：获取远程终端机构发送的待分析信息，所述待分析信息包括：眼底图像和个人资料；对所述待分析信息进行预判读，判断所述待分析信息是否合格；若所述待分析信息合格，对所述待分析信息提取特征数据，并形成结构化的量化指标；根据知识计算模型，对所述特征数据和量化指标进行整理与分析，获得分析结论；将所述待分析信息、特征数据、量化指标和分析结论存放至所述预先设计的数据库。
进一步的，所述“对所述待分析信息进行预判读，判断所述待分析信息是否合格”，还包括步骤：所述预判读包括：所述眼底图像是否确为眼底图像、所述眼底图像结构是否完整、所述眼底图像是否清晰和所述眼底图像是否可用中的一种或多种；若待分析信息合格，则返回相关合格信息至远程终端机构；若待分析信息不合格，则返回相关不合格信息至远程终端机构，所述相关不合格信息用以提示：远程终端机构重新采集待分析信息。
进一步的，所述“对所述待分析信息进行预判读，判断所述待分析信息是否合格”，还包括步骤：根据预设规则，在返回预判读结果至远程终端机构前，所述远程终端机构发出提示信息用以提示用户不要离开所述远程终端机构，直至返回待分析信息合格的提示信息。
进一步的，所述“对所述待分析信息进行预判读，判断所述待分析信息是否合格”，还包括步骤：若待分析信息合格，返回相关合格信息至远程终端机构；所述远程终端机构获取所述相关合格信息，并根据预设规则告知用户 是否需要继续等待直至所述分析结论出来。
进一步的，所述“所述眼底图像结构是否完整”，还包括步骤：对所述眼底图像的视盘与黄斑进行识别和标定，根据所述识别结果判断所述眼底图像是否包含视盘和黄斑，若所述眼底图像包含视盘和黄斑，根据所述标定结果判断所述视盘和黄斑是否在眼底图像的预设区间内，若所述视盘和黄斑在眼底图像的预设区间内，则所述眼底图像结构完整。
进一步的，所述“对所述眼底图像进行提取特征数据，并形成结构化的量化指标”，还包括步骤：根据标定好的视盘和黄斑，计算所述视盘颞侧与黄斑中心凹的量化参数。
为解决上述技术问题，还提供了一种远程眼底筛查的健康大数据服务***，具体技术方案如下：
一种基于远程眼底筛查的健康大数据服务***，包括：眼底图像采集模块和远程分析中心模块；所述眼底图像采集模块连接所述远程分析中心模块；所述眼底图像采集模块用于：获取待分析信息，所述待检信息包括：眼底图像和个人资料，并发送所述待分析信息至远程分析中心模块；所述远程分析中心模块用于：接收所述待分析信息，并对所述待分析信息进行预判读，判断所述待分析信息是否合格；若所述待分析信息合格，对所述待分析信息提取特征数据，并形成结构化的量化指标；根据知识计算模型，对所述特征数据和量化指标进行整理与分析，获得分析结论；将所述待分析信息、特征数据、量化指标和分析结论存放至所述预先设计的数据库。
进一步的，所述预判读包括：所述眼底图像是否确为眼底图像、所述眼底图像结构是否完整、所述眼底图像是否清晰和所述眼底图像是否可用中的一种或多种；所述远程分析中心模块还用于：若待分析信息合格，则返回相关合格信息至眼底图像采集模块；若待分析信息不合格，则返回相关不合格信息至眼底图像采集模块，所述相关不合格信息用以提示：眼底图像采集模块重新采集合格的待分析信息。
进一步的，所述眼底图像采集模块还用于：根据预设规则，在返回预判读结果至眼底图像采集模块前，发出提示信息用以提示用户不要离开眼底图像采集模块，直至返回待分析信息合格的提示信息。
进一步的，所述远程分析中心模块还用于：若待分析信息合格，返回相关合格信息至眼底图像采集模块；所述眼底图像采集模块还用于：获取所述相关合格信息，并根据预设规则告知用户是否需要继续等待直至所述分析结论出来。
本发明的有益效果是：通过获取远程终端机构发送的待分析信息，所述待检信息包括：眼底图像和个人资料，对所述待分析信息进行预判读，判断所述待分析信息是否合格，并构成一个完整闭环的质量保证体系，这一点十分重要，使得***能够使获得的每一个待分析信息，都是充分可用的，保证了可靠的用户信息的获得，增强了用户体验，有助于最终形成一个可分析、可更新的大数据信息库；若所述待分析信息合格，对所述待分析信息提取特征数据，并形成结构化的量化指标；将所述特征数据和所述量化指标存放至预先设计的数据库；根据知识计算模型，对所述特征数据和量化指标进行整理与分析，获得分析结论；将所述待分析信息、特征数据、量化指标和分析结论存放至所述预先设计的数据库。通过以上步骤使得无论是什么型号的眼底相机、什么样的工作模式、最后获得的眼底图像经过处理之后，都可以获得可比性统一的量化指标、特征数据，且将待分析信息、量化指标、特征数据、分析结论存放至预先设计好的数据库，建立起整个大数据服务平台，将大大地有利于辅助医生进行疾病判读等等。
进一步的，对待分析信息进行预判读，可百分百保证到最后用于提取特征数据的待分析信息是可用的，对于用户而言：避免了远程分析中心后期才发现待分析信息不可用，用户又要重新跑一趟，给用户很不好的体验，也浪费用户的时间；对于远程分析中心而言：可用的待分析信息不仅确保了诊断结果的稳定性、准确性，且提高诊断效率，避免无用的返工时间。
进一步的，在待分析信息合格之前，可根据预设规则让远程终端机构告知用户暂时不要离开，直至返回待分析信息合格的提示信息后，方可离开，整个过程避免了待分析信息不合格，用户却已经离开的情况，给用户造成不好的体验。
进一步的，若眼底图像合格，对所述眼底图像提取特征数据，并形成结构化的量化指标。即：根据标定好的视盘和黄斑，计算所述视盘颞侧与黄斑中心凹的量化参数。因正常人这二者的绝对距离值是差不多一样的，因此再根据已得到的视盘颞侧到黄斑中心凹的绝对距离以及视盘直径，得到后续量化分析的参数，将获得的数据从绝对表示方式转换为相对表示方式，通过归一化处理，形成有意义的、可比较的数据。确保了不同来源的眼底图像，均可形成有意义、可比较的量化指标，做到所有的眼底图像基本可比较。
附图说明
图1为具体实施方式所述一种基于远程眼底筛查的健康大数据服务方法的流程图；
图2为具体实施方式所述一种基于远程眼底筛查的健康大数据服务***的模块图。
附图标记说明：
200、基于远程眼底筛查的健康大数据服务***；
201、眼底图像采集模块；
202、远程分析中心模块。
具体实施方式
为详细说明技术方案的技术内容、构造特征、所实现目的及效果，以下结合具体实施例并配合附图详予说明。
请参阅图1，在本实施方式中，一种基于远程眼底筛查的健康大数据服务 方法中的全部或部分步骤可以通过程序来指令相关的硬件来完成，所述的程序可以存储于计算机设备可读取的存储介质中，用于执行下述各实施方式所述的全部或部分步骤。所述计算机设备，包括但不限于：个人计算机、服务器、通用计算机、专用计算机、网络设备、智能移动终端、智能家居设备、穿戴式智能设备等；所述的存储介质，包括但不限于：RAM、ROM、移动硬盘、网络服务器存储、网络云存储等。
在本实施方式中，一种基于远程眼底筛查的健康大数据服务方法的具体实施如下：
步骤S101：获取远程终端机构发送的待分析信息，所述待分析信息包括：眼底图像和个人资料。可采用如下方式：在远程终端机构设置有眼底图像采集终端和任意可接受发送信息的计算机设备，如PC，通过眼底图像采集终端采集眼底图像，再把眼底图像传输到PC上，同样将个人资料输入到PC中，由PC统一将眼底图像和个人资料发送给远程分析中心。
在其它实施方式中，亦可以是远程终端机构(如：底层医疗诊所)通过眼底照相机采集用户A的眼底图像，眼底照相机通过usb线将所述眼底图像传输到终端应用机构的电脑上，同时也可以将个人信息或个人资料输入电脑，电脑把这些信息发送至远程判读中心。
在本实施方式中，所述个人资料或问诊资料包括：姓名、身份证、身高、体重、腰围、家族遗传病史、用药情况、血糖、血压、视力情况、运动情况、饮食情况、生活习惯和是否吸烟喝酒中的一种或多种。
获取好待分析信息后，执行步骤S102：对所述待分析信息进行预判读，判断所述待分析信息是否合格。可采用如下方式：所述预判读包括：所述眼底图像是否确为眼底图像、所述眼底图像结构是否完整、所述眼底图像是否清晰和所述眼底图像是否可用中的一种或多种。
可采用如下方式：获取质检人员输入的预判读信息；结合所述质检人员输入的预判读信息及眼底图像自动分析的预判读结果，判断眼底图像是否合 格；所述输入信息包括：眼底图像质量等级。对所述眼底图像可对其进行自动分析，如：利用事先收集到的经专业医生进行质量等级划分的图像，训练SVM模型，使得该模型可以进行图像质量等级分类。这样，当基层传来一幅眼底图像后，一方面利用训练好的svm模型对眼底图像进行判断，另一方面亦获取质检人员(如：专业眼科医生)输入的对眼底图像质量等级判别的信息。结合以上二者对所述眼底图像进行预判读，通过二者的结合，人工辅助判断可避免机器自动分析的一些错误，而通过机器自动分析又可大大减轻人工识别的工作量与繁琐度，其中二者的结合可具体应用在：利用svm模型并识别为不合格而的眼底图像再经人工确认是否真不合格，避免误判，从而确保眼底图像最后百分百可用。
在其它实施方式中，亦可以是全自动地眼底图像自动分析，无需质检人员参与把关。
进一步的，对接收的眼底图像，可先判断所述眼底图像是否确为眼底图像，若所述眼底图像不是眼底图像，即：可能是远程终端机构人员恶作剧或选择发送眼底图像时操作错误等等引起的，则直接结束本次判读，并返回对应的提示信息给远程终端机构，告知其所采集到的眼底图像非眼底图像，用户相应地再进行眼底图像采集操作。若所述眼底图像是眼底图像，则判别所述眼底图像结构是否完整，可采用如下方式：对所述眼底图像的视盘与黄斑进行识别和标定，根据所述识别结果判断所述眼底图像是否包含视盘和黄斑，若所述眼底图像包含视盘和黄斑，根据所述标定结果判断所述视盘和黄斑是否在眼底图像的预设区间内，若所述视盘和黄斑在眼底图像的预设区间内，则所述眼底图像结构完整。
具体如下：对待检查眼底图像进行绿色通道选择、中值滤波、有限对比度增强和灰度的归一化处理。通过对眼底图像预处理，可去除眼底图像中多余的背景，有效去噪，更有利于后续眼底图像分析的进行。具体如下：任意一幅彩色眼底图像中，蓝色通道下噪声较多，有用的信息基本丢失，红色通 道下两斑较为突出，而暗色的血管、微血管瘤等信息丢失较多，故在本实施方式中对待检查的彩色眼底图像进行绿色通道选择，最大程度地保留、突出眼底血管。为去除噪声，且能较好地保留边界信息，本实施方式中对绿色通道下的眼底图像进行中值滤波，实现去噪；为能获得更好的血管提取的效果，对已经去噪的图像进行对比度增强。为避免图像增强后出现过亮的情况，本实施方式中采用有限的对比度增强方法CLAHE。最后，进行归一化处理，使得一幅图像中所有像素点的像素值均落在0-1之间。
对预处理后的眼底图像通过大津算法提取二值化血管图，并通过形态学方法对所述二值化血管图进行腐蚀得到主血管。可采用如下方式：对预处理后的眼底图像通过大津算法计算阈值，并根据以下公式将灰度值大于阈值的像素认定为血管；
Figure PCTCN2018124488-appb-000001
并根据视盘直径为图像宽度的1/8-1/5，及主血管的宽度为视盘直径的1/4构造结构元素，利用所述结构元素对已提取的血管进行腐蚀操作，去除细小血管，得到主血管。得到主血管后，对主血管进行抛物线拟合计算，根据计算结果定位视盘中心。可采用如下方式：以眼底图像左上角为原点，水平方向为X轴，垂直方向为Y轴，建立坐标系；将所述主血管中的各个像素点映射为所述坐标系的坐标；
如以下公式所示，根据最小二乘法对主血管进行抛物线拟合，确定抛物线的参数，并计算抛物线的顶点，
f(x)＝ax 2+bx+c
Figure PCTCN2018124488-appb-000002
判断所述抛物线顶点是否落在原眼底图像中，若所述抛物线顶点落在原眼底图像中，则定义所述抛物线顶点为视盘中心。基于外观特征与结构特征的黄斑定位：依据到黄斑与视盘间的位置关系，首先在已确定的视盘中心基 础上，进一步缩小中心凹的搜索范围。在一种优选方式下，由于黄斑中心凹与视盘中心的距离一般在2倍到3倍的视盘直径大小，以视盘中心为圆心，构造出环形掩模，将其定义为中心凹搜索范围；接着在搜索范围区域内，根据中心凹亮度最低的特点，完成中心凹的定位。在一种优选方式下，采用基于区域间亮度对比的快速搜索方式，确定中心凹的位置；最后根据亮度信息，以中心凹为圆心，圆形拟合黄斑区域。
在确定黄斑与视盘中心后，进行眼底图像结构完整性判定。根据表1所示的判决条件，满足的条件的图像即为完整性合格的图像。其中，Dod为视盘直径。
Figure PCTCN2018124488-appb-000003
表1
若所述眼底图像是眼底图像，则判别所述眼底图像结构是否清晰，可采用如下方式：判断所述眼底图像的视盘表面小血管及后极部视网膜神经纤维层是否可辨，若所述眼底图像的视盘表面小血管及后极部视网膜神经纤维层可辨，则所述眼底图像清晰度合格。具体如下：
a、根据已识别的视盘中心、黄斑中心凹，确定以视盘为圆心，1.5倍视盘直径范围的区域为感兴趣区域1，以黄斑中心凹为圆心，1倍视盘直径范围的区域为感兴趣区域2；
b、基于已确定的感兴趣区域1和区域2，选择某一清晰度评价算子，计算清晰度评价值，完成清晰度评价。
步骤S103：若所述待分析信息合格，对所述待分析信息提取特征数据，并形成结构化的量化指标。具体如下：根据标定好的视盘和黄斑，计算所述 视盘颞侧与黄斑中心凹的量化参数。根据视盘中心坐标和视盘半径，计算视盘颞侧坐标(ODX,ODY)；根据视盘颞侧坐标和黄斑中心凹坐标，计算视盘颞侧与黄斑中心凹的绝对距离，按以下公式计算该眼底图像中两者的欧式距离，作为本图像中视盘中心到黄斑中心凹的绝对距离；
Figure PCTCN2018124488-appb-000004
其中，所有坐标值均以眼底图像左上角像素为原点。
c、于黄斑中心凹一般距视盘颞侧缘约3mm，因此根据已得到的视盘颞侧到黄斑中心凹的绝对距离以及视盘直径，根据以下公式得到后续量化分析的标准d：
Figure PCTCN2018124488-appb-000005
在本实施方式中，以d为标尺，将获得的数据从绝对表示方式转换为相对表示方式，通过此归一化处理，形成有意义的、可比较的数据。
在本实施方式中，若已检测出硬性渗出，并已计算每个硬性渗出到黄斑中心凹的欧式距离Di。此时可根据公式1进行归一化处理。在此基础上，得到本眼底图像中，硬性渗出到黄斑中心凹的标准的最小距离。
Figure PCTCN2018124488-appb-000006
步骤S104：根据知识计算模型，对所述特征数据和量化指标进行整理与分析，获得分析结论。步骤S105：将所述待分析信息、特征数据、量化指标和分析结论存放至所述预先设计的数据库。
可采用如下方式：今天，用于糖尿病视网膜病变(DR)筛查的眼底照相机技术已经成熟，而且，DR的筛查有着与糖尿病病情相关并且能够指导治疗的防治指南和诊断标准；眼睛是人体全身唯一不用手术就能直接看到血管和神经的部位。医学证据表明：视网膜与脑部的循环***具有相近的解剖、生理和胚胎发育等特征。因此，通过眼底血管可以了解全身尤其脑部动脉和全身 中小动脉病变程度；根据我国高血压防治指南，视网膜动脉病变可反映小血管病变情况，如果我们能够通过眼底图象的定期筛查和比对，从中找出能够进行定量分析的关键方法，就能对不同人或同一个人不同时间或不同设备上采集的眼底图象的视网膜病变特征或血管改变情况进行定量分析、统计或比较，形成结构化的健康数据，通过所述“知识计算模型”以及依托所述“知识计算模型”建立起来的“疾病预警和健康评估引擎”就能对糖尿病视网膜病变、糖尿病肾病、高血压、脑卒中等疾病进行预警或特异性筛查，尤其是根据糖网患者如果不加以适当的生活方式干预基础治疗及药物治疗，其眼底视网膜病变特征或病情就一定会不断发展这一重要特点，建立或依托所述“知识计算模型”提供一种眼底图像血管特征的统计、计算和分析方法，这对于及时发现眼底视网膜病变和眼底血管改变等特征信息，提供辅助诊断信息或健康管理和健康服务建议，开展健康大数据服务，都具有十分积极的意义。
因此，根据所述用户眼底图像特征提取的特征和必要的个人资料，形成结构化的量化指标和所述深度专业设计的“知识库”，其中的所述量化指标包括：既往病史与身高、体重、腰围、日常生活运动饮食习惯、糖尿病史及先前治疗情况、高血压病史及先前治疗情况等健康情况以及家族遗传史、生活习惯等个人信息或问诊资料，微血管瘤、出血点、硬性渗出、棉絮斑等的个数、面积和位置相关的DR判读结果，加上是否出现新增生血管、是否出现黄斑水肿、感兴趣区域内动静脉比、动脉管径缩窄、是否动静脉交叉压痕及位置记录、金丝或银丝动脉等的情况、一种或多种血管改变及神经纤维层的变化情况等，就能依托所述“知识计算模型”，提供一种眼底视网膜病变、血管改变等特征的统计、计算和分析方法，即“疾病预警和健康评估引擎”。
通过获取远程终端机构发送的待分析信息，所述待检信息包括：眼底图像和个人资料，对所述待分析信息进行预判读，判断所述待分析信息是否合格；若所述待分析信息合格，对所述待分析信息提取特征数据，并形成结构化的量化指标；将所述特征数据和所述量化指标存放至预先设计的数据库； 根据知识计算模型，对所述特征数据和量化指标进行整理与分析，获得分析结论；将所述待分析信息、特征数据、量化指标和分析结论存放至所述预先设计的数据库。通过以上步骤使得无论是什么型号的眼底相机、什么样的工作模式、最后获得的眼底图像经过处理之后，都可以获得可比性统一的量化指标、特征数据，且将待分析信息、量化指标、特征数据、分析结论存放至预先设计好的数据库，建立起整个大数据服务平台，将大大地有利于辅助医生进行疾病判读等等。进一步的，对待分析信息进行预判读，可百分百保证到最后用于提取特征数据的待分析信息是可用的，对于用户而言：避免了远程分析中心后期才发现待分析信息不可用，用户又要重新跑一趟，给用户很不好的体验，也浪费用户的时间；对于远程分析中心而言：可用的待分析信息不仅确保了诊断结果的稳定性、准确性，且提高诊断效率，避免无用的返工时间。进一步的，在待分析信息合格之前，可根据预设规则让远程终端机构告知用户暂时不要离开，直至返回待分析信息合格的提示信息后，方可离开，整个过程避免了待分析信息不合格，用户却已经离开的情况，给用户造成不好的体验。进一步的，若眼底图像合格，对所述眼底图像提取特征数据，并形成结构化的量化指标。即：根据标定好的视盘和黄斑，计算所述视盘颞侧与黄斑中心凹的量化参数。因正常人这二者的绝对距离值是差不多一样的，因此再根据已得到的视盘颞侧到黄斑中心凹的绝对距离以及视盘直径，得到后续量化分析的参数，将获得的数据从绝对表示方式转换为相对表示方式，通过归一化处理，形成有意义的、可比较的数据。确保了不同来源的眼底图像，均可形成有意义、可比较的量化指标，做到所有的眼底图像基本可比较。
需要说明的是，在其它实施方式中，亦可以手动完成视盘与黄斑的定位。
在本实施方式中，在将“待分析信息发送至远程分析中心”步骤前，还可以包括步骤：远程终端机构设置有特定软件可用于对所述眼底图像和个人资料进行本地预判读，若预判读结果合格，则发出对应提示信息，用以提示：用户可以离开或继续在附近休息等待远程分析中心的分析结果。
进一步的，所述“对所述待分析信息进行预判读，判断所述待分析信息是否合格”，还包括步骤：根据预设规则，在返回预判读结果至远程终端机构前，所述远程终端机构发出提示信息用以提示用户不要离开所述远程终端机构，直至返回待分析信息合格的提示信息。可采用如下方式：若所述远程终端机构未设置有软件可用于对所述眼底图像和个人资料本地预判读，则根据预设规则(即：远程终端机构是否为所述远程分析中心的协议服务或紧密关联、不需要购买特定软件时，根据流程或质量控制体系上的约定)，在返回预判读结果至远程终端机构前，所述远程终端机构发出提示信息用以提示用户不要离开所述远程终端机构，直至返回待分析信息合格的提示信息，亦可以让所述用户继续在附近休息等待所述远程判读会诊中心的远程判读会诊结果。
进一步的，所述“对所述待分析信息进行预判读，判断所述待分析信息是否合格”，还包括步骤：若待分析信息合格，返回相关合格信息至远程终端机构；所述远程终端机构获取所述相关合格信息，并根据预设规则告知用户是否需要继续等待直至所述分析结论出来。可采用如下方式：远程终端结构接收到待分析信息合格的提示后，可根据实际情况告知用户是否需要继续等待直至分析结论出来。
请参阅图2，在本实施方式中，所述眼底图像采集模块201至少包括：眼底图像采集相机和电脑；所述远程分析中心模块202可为存储设备。在该存储设备上安装有对应的远程分析中心APP，或直接打开对应的远程分析中心网页，即可对眼底图像采集模块201传送过来的待分析信息进行处理。一种基于远程眼底筛查的健康大数据服务***200的具体实施方式如下：
一种基于远程眼底筛查的健康大数据服务***200，包括：眼底图像采集模块201和远程分析中心模块202；所述眼底图像采集模块201连接所述远程分析中心模块202；所述眼底图像采集模块201用于：获取待分析信息，所述待检信息包括：眼底图像和个人资料，并发送所述待分析信息至远程分析中 心模块202；所述远程分析中心模块202用于：接收所述待分析信息，并对所述待分析信息进行预判读，判断所述待分析信息是否合格；若所述待分析信息合格，对所述待分析信息提取特征数据，并形成结构化的量化指标；根据知识计算模型，对所述特征数据和量化指标进行整理与分析，获得分析结论；将所述待分析信息、特征数据、量化指标和分析结论存放至所述预先设计的数据库。进一步的，所述预判读包括：所述眼底图像是否确为眼底图像、所述眼底图像结构是否完整、所述眼底图像是否清晰和所述眼底图像是否可用中的一种或多种；所述远程分析中心模块202还用于：若待分析信息合格，则返回相关合格信息至眼底图像采集模块201；若待分析信息不合格，则返回相关不合格信息至眼底图像采集模块201，所述相关不合格信息用以提示：眼底图像采集模块201重新采集合格的待分析信息。
进一步的，所述眼底图像采集模块201还用于：根据预设规则，在返回预判读结果至眼底图像采集模块201前，发出提示信息用以提示用户不要离开眼底图像采集模块201，直至返回待分析信息合格的提示信息。
进一步的，所述远程分析中心模块202还用于：若待分析信息合格，返回相关合格信息至眼底图像采集模块201；所述眼底图像采集模块201还用于：获取所述相关合格信息，并根据预设规则告知用户是否需要继续等待直至所述分析结论出来。
一种基于远程眼底筛查的健康大数据服务***200通过眼底图像采集模块201获取待分析信息，所述待检信息包括：眼底图像和个人资料，远程分析中心模块202对待分析信息进行预判读，判断所述待分析信息是否合格；若所述待分析信息合格，对所述待分析信息提取特征数据，并形成结构化的量化指标；根据知识计算模型，对所述特征数据和量化指标进行整理与分析，获得分析结论；将所述待分析信息、特征数据、量化指标和分析结论存放至所述预先设计的数据库。通过以上模块功能的实现使得无论是什么型号的眼底相机、什么样的工作模式、最后获得的眼底图像经过处理之后，都可以获 得可比性统一的量化指标、特征数据，且将待分析信息、量化指标、特征数据、分析结论存放至预先设计好的数据库，建立起整个大数据服务平台，将大大地有利于辅助医生进行疾病判读等等。
进一步的，对待分析信息进行预判读，可百分百保证到最后用于提取特征数据的待分析信息是可用的，对于用户而言：避免了远程分析中心后期才发现待分析信息不可用，用户又要重新跑一趟，给用户很不好的体验，也浪费用户的时间；对于远程分析中心而言：可用的待分析信息不仅确保了诊断结果的稳定性、准确性，且提高诊断效率，避免无用的返工时间。
进一步的，在待分析信息合格之前，可根据预设规则让远程终端机构告知用户暂时不要离开，直至返回待分析信息合格的提示信息后，方可离开，整个过程避免了待分析信息不合格，用户却已经离开的情况，给用户造成不好的体验。
进一步的，若眼底图像合格，对所述眼底图像提取特征数据，并形成结构化的量化指标。即：根据标定好的视盘和黄斑，计算所述视盘颞侧与黄斑中心凹的量化参数。因正常人这二者的绝对距离值是差不多一样的，因此再根据已得到的视盘颞侧到黄斑中心凹的绝对距离以及视盘直径，得到后续量化分析的参数，将获得的数据从绝对表示方式转换为相对表示方式，通过归一化处理，形成有意义的、可比较的数据。确保了不同来源的眼底图像，均可形成有意义、可比较的量化指标，做到所有的眼底图像基本可比较。
需要说明的是，尽管在本文中已经对上述各实施例进行了描述，但并非因此限制本发明的专利保护范围。因此，基于本发明的创新理念，对本文所述实施例进行的变更和修改，或利用本发明说明书及附图内容所作的等效结构或等效流程变换，直接或间接地将以上技术方案运用在其他相关的技术领域，均包括在本发明的专利保护范围之内。

Claims (10)

  1. 一种基于远程眼底筛查的健康大数据服务方法，其特征在于，包括步骤：获取远程终端机构发送的待分析信息，所述待分析信息包括：眼底图像和个人资料；对所述待分析信息进行预判读，判断所述待分析信息是否合格；若所述待分析信息合格，对所述待分析信息提取特征数据，并形成结构化的量化指标；根据知识计算模型，对所述特征数据和量化指标进行整理与分析，获得分析结论；将所述待分析信息、特征数据、量化指标和分析结论存放至所述预先设计的数据库。
  2. 根据权利要求1所述的一种基于远程眼底筛查的健康大数据服务方法，其特征在于，所述“对所述待分析信息进行预判读，判断所述待分析信息是否合格”，还包括步骤：所述预判读包括：所述眼底图像是否确为眼底图像、所述眼底图像结构是否完整、所述眼底图像是否清晰和所述眼底图像是否可用中的一种或多种；若待分析信息合格，则返回相关合格信息至远程终端机构；若待分析信息不合格，则返回相关不合格信息至远程终端机构，所述相关不合格信息用以提示：远程终端机构重新采集待分析信息。
  3. 根据权利要求1所述的一种基于远程眼底筛查的健康大数据服务方法，其特征在于，所述“对所述待分析信息进行预判读，判断所述待分析信息是否合格”，还包括步骤：根据预设规则，在返回预判读结果至远程终端机构前，所述远程终端机构发出提示信息用以提示用户不要离开所述远程终端机构，直至返回待分析信息合格的提示信息。
  4. 根据权利要求1所述的一种基于远程眼底筛查的健康大数据服务方法，其特征在于，所述“对所述待分析信息进行预判读，判断所述待分析信息是否合格”，还包括步骤：若待分析信息合格，返回相关合格信息至远程终端机构；所述远程终端机构获取所述相关合格信息，并根据预设规则告知用户是否需要继续等待直至所述分析结论出来。
  5. 根据权利要求2所述的一种基于远程眼底筛查的健康大数据服务方法，其特征在于，
    所述“所述眼底图像结构是否完整”，还包括步骤：对所述眼底图像的视盘与黄斑进行识别和标定，根据所述识别结果判断所述眼底图像是否包含视盘和黄斑，若所述眼底图像包含视盘和黄斑，根据所述标定结果判断所述视盘和黄斑是否在眼底图像的预设区间内，若所述视盘和黄斑在眼底图像的预设区间内，则所述眼底图像结构完整。
  6. 根据权利要求5所述的一种基于远程眼底筛查的健康大数据服务方法，其特征在于，所述“对所述眼底图像进行提取特征数据，并形成结构化的量化指标”，还包括步骤：根据标定好的视盘和黄斑，计算所述视盘颞侧与黄斑中心凹的量化参数。
  7. 一种基于远程眼底筛查的健康大数据服务***，其特征在于，包括：眼底图像采集模块和远程分析中心模块；所述眼底图像采集模块连接所述远程分析中心模块；所述眼底图像采集模块用于：获取待分析信息，所述待检信息包括：眼底图像和个人资料，并发送所述待分析信息至远程分析中心模块；所述远程分析中心模块用于：接收所述待分析信息，并对所述待分析信息进行预判读，判断所述待分析信息是否合格；若所述待分析信息合格，对所述待分析信息提取特征数据，并形成结构化的量化指标；根据知识计算模型，对所述特征数据和量化指标进行整理与分析，获得分析结论；将所述待分析信息、特征数据、量化指标和分析结论存放至所述预先设计的数据库。
  8. 根据权利要求7所述的一种基于远程眼底筛查的健康大数据服务***，其特征在于，所述预判读包括：所述眼底图像是否确为眼底图像、所述眼底图像结构是否完整、所述眼底图像是否清晰和所述眼底图像是否可用中的一种或多种；所述远程分析中心模块还用于：若待分析信息合格，则返回相关合格信息至眼底图像采集模块；若待分析信息不合格，则返回相关不合格信息至眼底图像采集模块，所述相关不合格信息用以提示：眼底图像采集模块重新采集合格的待分析信息。
  9. 根据权利要求7所述的一种基于远程眼底筛查的健康大数据服务***， 其特征在于，所述眼底图像采集模块还用于：根据预设规则，在返回预判读结果至眼底图像采集模块前，发出提示信息用以提示用户不要离开眼底图像采集模块，直至返回待分析信息合格的提示信息。
  10. 根据权利要求7所述的一种基于远程眼底筛查的健康大数据服务***，其特征在于，所述远程分析中心模块还用于：若待分析信息合格，返回相关合格信息至眼底图像采集模块；所述眼底图像采集模块还用于：获取所述相关合格信息，并根据预设规则告知用户是否需要继续等待直至所述分析结论出来。
PCT/CN2018/124488 2018-08-31 2018-12-27 一种基于远程眼底筛查的健康大数据服务方法和*** WO2020042500A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US17/271,611 US20210391056A1 (en) 2018-08-31 2018-12-27 Health big data service method and system based on remote fundus screening

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201811013848 2018-08-31
CN201811013848.2 2018-08-31

Publications (1)

Publication Number Publication Date
WO2020042500A1 true WO2020042500A1 (zh) 2020-03-05

Family

ID=69643878

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/124488 WO2020042500A1 (zh) 2018-08-31 2018-12-27 一种基于远程眼底筛查的健康大数据服务方法和***

Country Status (3)

Country Link
US (1) US20210391056A1 (zh)
CN (1) CN110875092B (zh)
WO (1) WO2020042500A1 (zh)

Families Citing this family (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111528790B (zh) * 2020-04-21 2023-08-04 张姬娟 一种取像装置、视力数据处理方法和眼底数据处理方法
CN112487012B (zh) * 2020-11-12 2022-05-27 苏州浪潮智能科技有限公司 一种基于数据筛选的部件参数获取方法及装置
CN112907548A (zh) * 2021-02-26 2021-06-04 依未科技(北京)有限公司 图像评估方法及装置、计算机可读存储介质及电子设备
EP4315020A1 (en) * 2021-03-30 2024-02-07 Jio Platforms Limited System and method of data ingestion and processing framework
CN113425248B (zh) * 2021-06-24 2024-03-08 平安科技（深圳）有限公司 医疗影像评估方法、装置、设备及计算机存储介质
CN114972462B (zh) * 2022-07-27 2023-08-15 北京鹰瞳科技发展股份有限公司 对眼底相机的工作距离对齐效果进行优化的方法及其相关产品
CN115423804B (zh) * 2022-11-02 2023-03-21 依未科技(北京)有限公司 影像标定方法及装置、影像处理方法
CN115579109A (zh) * 2022-11-24 2023-01-06 合肥心之声健康科技有限公司 医疗环境下心电图图像分析方法、装置和终端设备
CN115517686A (zh) * 2022-11-24 2022-12-27 合肥心之声健康科技有限公司 家庭环境心电图图像分析方法、装置、设备、介质和***

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150104087A1 (en) * 2013-10-10 2015-04-16 University Of Rochester Automated Fundus Image Field Detection and Quality Assessment
CN204698510U (zh) * 2015-06-02 2015-10-14 福州大学 图像质量保证的糖尿病性视网膜病变眼底筛查照相装置
CN105411525A (zh) * 2015-11-10 2016-03-23 广州河谷互动医疗科技有限公司 一种眼底照片图像智能获取识别***
CN107194179A (zh) * 2017-05-26 2017-09-22 苏州微清医疗器械有限公司 眼底相机及眼底图像拍摄方法
CN107423571A (zh) * 2017-05-04 2017-12-01 深圳硅基仿生科技有限公司 基于眼底图像的糖尿病视网膜病变识别***
CN108272434A (zh) * 2017-12-07 2018-07-13 江威 对眼底图像进行处理的方法及装置

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP5723093B2 (ja) * 2009-11-25 2015-05-27 キヤノン株式会社 画像処理装置、画像処理装置の制御方法及びプログラム
WO2015013632A1 (en) * 2013-07-26 2015-01-29 The Regents Of The University Of Michigan Automated measurement of changes in retinal, retinal pigment epithelial, or choroidal disease
CN107680683A (zh) * 2017-10-09 2018-02-09 上海睦清视觉科技有限公司 一种ai眼部健康评估方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150104087A1 (en) * 2013-10-10 2015-04-16 University Of Rochester Automated Fundus Image Field Detection and Quality Assessment
CN204698510U (zh) * 2015-06-02 2015-10-14 福州大学 图像质量保证的糖尿病性视网膜病变眼底筛查照相装置
CN105411525A (zh) * 2015-11-10 2016-03-23 广州河谷互动医疗科技有限公司 一种眼底照片图像智能获取识别***
CN107423571A (zh) * 2017-05-04 2017-12-01 深圳硅基仿生科技有限公司 基于眼底图像的糖尿病视网膜病变识别***
CN107194179A (zh) * 2017-05-26 2017-09-22 苏州微清医疗器械有限公司 眼底相机及眼底图像拍摄方法
CN108272434A (zh) * 2017-12-07 2018-07-13 江威 对眼底图像进行处理的方法及装置

Also Published As

Publication number Publication date
US20210391056A1 (en) 2021-12-16
CN110875092B (zh) 2023-10-20
CN110875092A (zh) 2020-03-10

Similar Documents

Publication Publication Date Title
WO2020042500A1 (zh) 一种基于远程眼底筛查的健康大数据服务方法和***
CN109784337B (zh) 一种黄斑区识别方法、装置及计算机可读存储介质
US10441160B2 (en) Method and system for classifying optic nerve head
Patton et al. Retinal image analysis: concepts, applications and potential
Trucco et al. Validating retinal fundus image analysis algorithms: issues and a proposal
CN111222361B (zh) 一种高血压视网膜血管改变特征数据分析的方法和***
Niemeijer et al. Automated detection and differentiation of drusen, exudates, and cotton-wool spots in digital color fundus photographs for diabetic retinopathy diagnosis
AU2014271202B2 (en) A system and method for remote medical diagnosis
Xiong et al. An approach to evaluate blurriness in retinal images with vitreous opacity for cataract diagnosis
TWI719587B (zh) 眼底圖像定量分析的前置處理方法及存放裝置
WO2020042404A1 (zh) 用于远程眼底筛查的质量控制方法和***、及存储设备
CN111785363A (zh) 一种基于ai引导的慢性病辅助诊断***
CN106214120A (zh) 一种青光眼的早期筛查方法
CN113768461A (zh) 一种眼底图像分析方法、***和电子设备
CN111402184B (zh) 一种实现远程眼底筛查和健康服务的方法和***
Popovic et al. Fractal characterization of retinal microvascular network morphology during diabetic retinopathy progression
CN111292286B (zh) 糖网眼底特征数据变化的分析方法与***，及存储设备
CN111383775A (zh) 一种基于眼底筛查的会员用户匿名交流方法及存储设备
CN111435612B (zh) 一种移动医疗个性化健康服务的方法和***
CN116030042A (zh) 一种针对医生目诊的诊断装置、方法、设备及存储介质
CN111325631B (zh) 一种基于眼底图像的健康保险控费的方法和***
WO2024147161A1 (en) System and method for detecting retinopathy of prematurity
CN116157829A (zh) 用于使用光学相干体层摄影血管造影图像对视网膜微脉管***进行基于云的自动定量评估的方法和***
WO2024089642A1 (en) System and method for detecting glaucoma
TW202424993A (zh) 視網膜自動分層模型之建立方法、視網膜分層量化系統、眼部照護裝置、視網膜分層厚度及視網膜分層面積的檢測方法及神經退化性疾病評估和預測方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18931928

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18931928

Country of ref document: EP

Kind code of ref document: A1